A new web-based method for automatic selection of articles for systematic literature reviews

A systematic review about a particular subject provides the basis of knowledge for supporting a research. The increase of scientific information available and the easy access to such information in electronic databases have contributed to the rise of systematic review studies. However, one of the problems that arise with traditional research methods is the difficulty of reading all available articles. Thus, some items must be selected according to predefined selection rules. Nevertheless, the effectiveness of a systematic review is directly related to the relevance of the scientific papers selected according to the purpose of the particular study, among others factors. Therefore, there are several indicators that can be used to prioritize the articles that will set the articles portfolio of the review. The objective of this article is to introduce a method and a web system, implemented in R package, designed to apply automated filters to help in the selection of articles for systematic reviews. We propose two filters in the discussed method: the journal impact factor and the number of citations of the articles. The latter was analyzed by using Pareto rule. In the portfolio creation process, articles are first selected from a query performed in the electronic databases Scopus and Web of Science. Then, the proposed method can be automatically applied using the web system introduced here.